Problems opening already processed files

Hi all,
Sorry if this doesn’t make much sense, but I am not an expert in image processing. I’m having a bit of a perplexing problem.
I processed a set of images in Fiji and saved the final images as tiff files.
All of the images are 32 bit and have regions that after thresholding are set to NaN.
Now for the problems:
When I open these processed files using Fiji on the Mac I first processed them on, they open fine, with the NaN areas still set as NaN and the intensity range with the appropriate range.
When however I try to open these files using Fiji on a different Mac I usually end up with an image in which the NaN regions are set to huge negative values and the intensity range is also messed up. I haven’t found a quick work around either, other than re-processing the all the files (time consuming and shouldn’t be necessary!). The other weird thing is that with certain files the processing is kept intact! I compared the info of two of these images and the only difference was that the PhotometricInterpretation was different (the correct image had Black is Zero while the incorrect processed image had White is Zero).
Can anyone explain what might be going on?
I Have a feeling it might be because of different versions of Fiji being used (the files are processed in an older version of Fiji, which I can’t update as I don’t have admin rights, it’s my work computer), but then don’t know what I can do to fix this or if this even makes sense.

Thanks in advance!

Just to clarify, I have tried opening with Bio-formats plugin, drag-n-drop and even import as Raw and still doesn’t fix the problem.

Welcome to the forum, @Antonio_Kon!

That is a plausible theory.

You should not need admin rights to run ImageJ or Fiji. Just extract the application into the Applications folder of your home folder, and run from there.

Maybe this guide could help to track down exactly what the differences are.

It would help if you could share the exact workflow you use (ideally as a macro or script), so that we can attempt to reproduce.

1 Like

Thank you for your reply @ctrueden !
So the workflow for creating the mask after opening a 32-bit image is as follows:
Select image:

waitForUser("Threshold","set 50...500")
setAutoThreshold("Default dark");
setThreshold(50.0000, 491.0000);

I apply threshold but without setting 0 values to NaN.

I then re-select the image:

run("Divide...", "value=255.000");
setAutoThreshold("Default dark");
run("NaN Background");

I then save that file as a tiff as a mask
Then I use image calculator to multiply the mask with another 32-bit image:

imageCalculator("Multiply create 32-bit", "mask.tif","alpha.tif");

It’s a bit clunky but whenever I have tried to make a full script it never works with the necessary results, but I’m learning.

I unfortunately don’t have access to the work Mac (I am working off site for a few months) so won’t be able to test the settings issue. But the admin issue was a problem as nothing could be installed without IT…I could have asked them to update for me…but it seemed unnecessary as things worked fine.

Thanks for your help and suggestions.

Good day Antonio,

just curious, could you please explain the logic behind

waitForUser(“Threshold”,“set 50…500”)
setAutoThreshold(“Default dark”);
setThreshold(50.0000, 491.0000);



Hi @anon96376101,
I am acquiring fluorescence lifetime microscopy data and I acquire up to 500 photons per pixel, but in dimmer regions of the field of view I don’t reach as many photons. In order to make sure the data is representative of the actual result and not just due to background fluorescence I design a mask with an intensity threshold, with the minimum amount of photons set to 50.
As for the waitForUser command, I used this more for posterity in case any of my students or anyone else wanting to analyse the data wants to see the threshold I set in analysis.